Australia State of the Environment Report 2001 (Theme Report)
Lead Author: Professor Peter W. Newton, CSIRO Building, Construction and Engineering, Authors
Published by CSIRO on behalf of the Department of the Environment and Heritage, 2001
ISBN 0 643 06747 7
Liveability: human well-being (continued)
Social and economic well-being (continued)
A major contemporary issue is the differential levels of access of the population to new technologies such as personal computers and the Internet. While the rate of Internet connection is increasing, there is evidence of a digital divide. By November 1999, 3.2 million households in Australia had a home computer, and almost half of these had Internet access (ABS 2000i). This trend is growing rapidly although not uniformly (Table 36). The low level of connectedness to the Internet (approximately one-quarter of households) and the slower rate of increase for certain family types - especially single parents - and for low-income households in general, illustrates the nature of the 'digital divide'. There is also a significant gap between the metropolitan cities and rural and regional areas, with those in cities having higher levels of access. In an age where the ability to be able to tap into the Internet has become an important part of both human capital development and the operation of business-individual and corporate-individuals, families and firms who find it difficult or impossible to access these technologies and information networks will be at a significant disadvantage (Newton 1995).
|Households with home computer||3.2 million (47%)|
|Households with Internet access||1.6 million (25.1%)|
|Age of person frequently using the Internet (%)|
|Family type with Internet access (%)|
|Couples with no children||18.0|
|Couples with children||35.0|
|Income of households with Internet access (%)|
|$0-39, 000 pa||34.0|
|$40-79 000 pa||63.0|
|$80 000 and above||73.0|
|Home Internet access by region|
|Rest of Australia||17.0|
A challenging issue is how universal service obligations - guaranteeing a minimal level of service in people's access to telecommunications services - will continue to be implemented, and there is widespread concern over the regional differences in access to many services. There is a marked city focus in the private market-driven telecommunications sector. The big cities offer the advantage of having the highest demand, lowest marketing and servicing costs, and the highest anticipated revenue stream (Newton 1995, Morris 1999). The rural, regional and remote communities are disadvantaged in the availability and adoption of new services, being locations without high demand, but with high costs of supplying infrastructure and requiring heavy cross-subsidisation. All this reinforces the already strong agglomeration of economic activities, and especially in Sydney and Melbourne in the telecommunications-intensive activities, thus reinforcing both the urban and corporate hierarchy dominated by Australia's largest cities. Bryant (1999) notes how these problems extend beyond network access to include elements such as technical support, training and skills development, and so on-issues that are intended to be addressed via the federal 'Networking the Nation' initiative funded by the part sale of Telstra, Australia's largest telecommunications provider.
The ability of Australians to consume a range of goods and services, to provide a home, and to participate actively in society all rely to varying degrees on the ability to access a minimum standard of income.
An emerging issue over recent decades is the question of inequalities in income distribution and in people's access to material resources. This concern follows periods of real income growth for Australia's population as a whole and an equalising trend in the income distribution until the late 1970s. Over the past decade, however, studies have suggested that not all population groups within Australian society have shared equally in this period of economic and income growth (NATSEM 2000, Saunders 2000). There appears to be growing research evidence, however, to show that the rapid economic changes which occurred in the last decade or two of the 20th century have resulted in widening gaps between socio-economic groups in Australian society. The growth in high-paid jobs, together with a continuation of relatively high levels of unemployment among certain parts of the population, have contributed to this change.
The 1996 SoE report noted mixed findings in this regard, with some suggesting increases in inequalities (Saunders 1994), while others claimed that inequalities had in fact declined (Johnson et al. 1995). The main difference between these findings appears to be methodological differences, including a different definition of income. The 1996 SoE report also noted income differentials by gender and ethnic background.
The debate still continues, and it is fraught with difficulties, particularly about how income is measured-should it be gross income, not adjusted for size of income units, or an after-tax measure equalised for the composition of income units? Trends in the distribution of gross income will tend to reflect the operation of the market and private sector forces. Trends in the distribution of disposable income will tend to reflect the effects of government intervention (i.e. taxes and benefits). Different measures of income and different ways of analysing the distribution can influence, to a degree, the patterns that emerge.
Using the P90/P10 ratio, ABS (unpublished data, 2000) analysed recent trends (1994-2000) in income distribution. Using the gross income measure, Figure 42 shows that levels of income inequality increased between 1994-1995 and 1999-2000. In 1994, those at the top of the distribution had a gross income 7.9 times higher than those at the bottom and, at the end of the period, this ratio had risen to 8.5. The increase in inequality is less evident using the disposable income measure, since this reflects the effect of government policy towards income equalisation via taxes (of higher incomes) and benefits (to those on lower incomes).
Figure 42: P90/P10 income ratio among income units, Australia.
P90 is the income of the income unit at the 90th percentile of the distribution, and P10 is the income of the income unit at the 10th percentile of the distribution. The P90/P10 ratio measures how much more an income unit close to the top of the income distribution receives to an income unit close to the bottom. Various equivalence scales can be used. These include those known as the Henderson scales (developed for the 1975 Commission of Inquiry into Poverty in Australia) and the now frequently used OECD equivalence scales.
Source: ABS (2001)
The ABS has also calculated a Gini coefficient for gross income distribution in Australia. Between 1994-95 and 1999-2000 this coefficient increased marginally, from 0.443 to 0.448 (Figure 43). Again using ABS data from income surveys (which typically have sampling standard errors of around [+/-]2.8%) over the period from the mid-1980s to the late 1990s, it is evident that there was a widening of gross income inequality. For example, in 1984 the top 25% of income units earned 38.8% of gross income, while the bottom 25% earned 6.3% (Figure 44). By 1997-98, the top 25% accounted for almost 50% of gross income, while the bottom 25% earned only 3.8% of that income. When one takes a shorter and more recent time perspective of this changing gross income distribution, then the ABS data shows that, between 1994-95 and 1999-2000, the share of the lowest quintile for gross income had increased from 3.6% to 3.8%, while the share of the highest quintile increased its share of gross income from 47.9% to 48.5%. There has thus been a decline in the shares of the middle 60%; the second quintile's share declined from 9.3% to 9.0%; the third quintile's share declined from 15.2% to 15.0%; and the fourth quintile's share declined from 24.0% to 23.8%.
Figure 43: Distribution of income among income units in Australia: Gini coefficients.
Source: ABS Income Distribution Surveys
Figure 44: Share of income, gross weekly income quintiles, 1984 and 1998.
Source: ABS (1998 and various years)
Admittedly, these shifts showing a continuation of the 'hollowing out' of the middle income earners in Australia-something identified by researchers for almost two decades-are small shifts, and might reflect a decline in the rate of income inequality change occurring. The ABS, in its 1999-2000 Income Distribution bulletin (ABS 2000i), made the point that an analysis based on gross income takes no account of the distributional impact of the income tax system or differences in the composition of income units. If disposable income is used (deducting personal income tax and the Medicare levy from each income unit's gross cash income), the Gini coefficient in 1999-2000 is 0.396, compared to 0.448 for gross weekly income, and when the OECD equivalent adjustments are made, the Gini coefficient is further reduced.
These reductions in the Gini coefficient would be expected, as the tax and welfare systems are supposed to have an income redistributional effect in favour of lower income and disadvantaged family households. Indeed, on the disposable income distribution measure, the lowest income quintile's share was 4.7% in 1999-2000 and the share of the highest quintile was 44.3% (these shares being 7.1% and 38.5% respectively for the Henderson equivalent measure).
Nonetheless, the gross income distribution data displayed in Figure 44 indicate that, since the 1980s, there has been a shift in the distribution of gross income of Australians from the poor to the rich, while the ABS analysis of disposable income by the Henderson equivalent shows that income inequality is being offset by interventions through the tax and social security system which, according to the Gini coefficients for 1999-2000, are having a redistributional effect of between 5% and 13%.
Using data for an earlier period, Norris and McLean (1999) show a similar trend in the distribution of incomes. Dividing employees into three income categories, they showed that between 1975 and 1998 there was significant growth in both the high-paid category and low-paid category, and a general hollowing out of the middle income earners (Table 37).
Source: Norris and McClean(1999).
Further confirmation of these trends is provided by Saunders (2000) who showed that '... between 1985 and 1998, earnings at the [lowest] 10th percentile rose by less than the CPI (ie the level of real earnings declined). Earnings at the median and 90th percentile rose substantially faster than the CPI'.
But aggregate distribution of income (however measured) across the nation's income units tells us nothing about the marked geographical variations that occur-something that economists tend not to focus on because they take a non-spatial view of the world, whereas geographers take a spatial view of the world.
In relation to regional variations in income, a study by Lloyd et al. (2000, p.22) concluded that there is a 'large and growing gap between the incomes of those Australians living in the capital cities and those living in the rest of Australia. The incomes of metropolitan residents increased at about double the rate of those living in the major urban centres and regional and rural towns in the five years to 1996. However, people living in rural areas (not rural towns) enjoyed by far the strongest income increase between 1991 and 1996... The results indicate that "regional Australia" is not uniformly disadvantaged and not uniformly declining. The biggest losses appear to be to the residents of small rural towns rather than residents of rural areas' (Table 38).
|Class of settlementA||Average household income ($'000s) (in 1996 dollars)|
AMajor urban regions have populations over 100 000, regional towns 1000-100 000, and rural towns 200-999.
Source: Lloyd et al. (2000)
When Lloyd et al. (2000) analysed income distribution patterns and trends across local government areas in Australia in 1991 and 1996, they noted how 'spatial income inequality increased', as 'average household income grew strongly in the most affluent LGAs and declined in the poorest LGAs... ' This suggests a continuation of the process of increasing geographic polarisation of households within Australia's cities and towns first identified by Gregory and Hunter (1995).
A mapping of the index of economic resources developed by the ABS (1998c) further illustrates the uneven nature of socio-economic well-being across Australia's human settlements. Known as the SEIFA (Socio Economic Index For Areas) index, it provides a profile of the economic resources of families for defined areas. The index is produced using 1996 census data that include household income and expenditure (income, rent) as well as non-income assets such as dwelling size and number of cars. It is clear that residents of big cities and other urban areas have a larger share of economic resources than do residents in other human settlements, although people in remote centres, are also 'better off', reflecting the location of significant mining activities (Figure 45). Not surprisingly, metropolitan Sydney,Melbourne and Perth, together with Canberra, recorded a higher level of economic resources than other capital cities, a fact reflecting the performance of these cities in the national and international economy (Figure 46). These patterns reflect many factors, including the varying range of occupations and industries found in centres of different size, and differences in the number of wage earners per household-the proportion of households with two wage earners or more is much higher in the major urban centres and rural balance than elsewhere (ABS 1998a). A national study by Baum et al. (1999) on community opportunity and vulnerability in Australia presented a similar picture.
Figure 45: Distribution of economic resources across Australia, 1996. [HS Indicator 3.11]
Source: ABS (1996c)
Figure 46: Distribution of economic resources in capital cities, 1996. [HS Indicator 3.11]
Source: ABS (1996c)